Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Feature selection by Moran'I score #110

Closed
wants to merge 4 commits into from
Closed

Feature selection by Moran'I score #110

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
e907bbd
Feature selection by Moran'I score
Dec 13, 2023
fa9f6f1
Update spatially_variable_genes_scanpy.py
Qirongmao97 Dec 13, 2023
6fef445
Removed if args.n_top is None
Qirongmao97 Dec 13, 2023
7fafb20
Aligning with the output of /preprocessing/neibours
Qirongmao97 Dec 13, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 1 addition & 9 deletions preprocessing/feature_selection_MoranI/spatially_variable_genes_scanpy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,7 @@
"-o", "--observations", help="Path to observations (as tsv).", required=True
)
parser.add_argument(
"-n", "--n_top_genes", help="Number of genes to keep (Default:3000).", required=False, type=int
)
"-n", "--n_top_genes", help="Number of genes to keep (Default:3000).", required=False, type=int, default=3000)

parser.add_argument("-d", "--out_dir", help="Output directory.", required=True)

Expand Down Expand Up @@ -51,8 +50,6 @@

if args.n_top_genes is not None:
n_top_genes = int(args.n_top_genes)
Qirongmao97 marked this conversation as resolved.
Show resolved Hide resolved
else:
n_top_genes= 3000 ### Default: 3000 genes

niklasmueboe marked this conversation as resolved.
Show resolved Hide resolved

if args.config is not None:
Expand Down Expand Up @@ -98,11 +95,6 @@ def get_anndata(args):

features_df = adata.var.copy()

## Log Normalization

sc.pp.normalize_total(adata)
sc.pp.log1p(adata)

## Compute the graph based from Delaunay triangulation
sq.gr.spatial_neighbors(adata,coord_type="generic",delaunay= True)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wondering if that should be a commandline arg rather than being calculated on the fly. Otherwise we will have situation later where the neighborhood structure used in the method might be different than the one used for feature selection which could be awkward.
@naveedishaque

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For Visium we could just use the default?

For others there are parameters like:
n_neighs ([int])
radius ([int])
delaunay ([bool])

Reference: https://squidpy.readthedocs.io/en/stable/api/squidpy.gr.spatial_neighbors.html

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wondering if that should be a commandline arg rather than being calculated on the fly. Otherwise we will have situation later where the neighborhood structure used in the method might be different than the one used for feature selection which could be awkward.

Oh, this is interesting. We want to make sure that the transformations and neighborhood structure is controlled outside, so technically this would need to import the log1p data and import the neighborhood graph.

Comments:

  • branch name change to "preproc_MoranI_scanpy_Qirongmao"

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wondering if that should be a commandline arg rather than being calculated on the fly. Otherwise we will have situation later where the neighborhood structure used in the method might be different than the one used for feature selection which could be awkward.

Oh, this is interesting. We want to make sure that the transformations and neighborhood structure is controlled outside, so technically this would need to import the log1p data and import the neighborhood graph.

Comments:

  • branch name change to "preproc_MoranI_scanpy_Qirongmao"

Then should we need to make a separate pipeline for generating the neighborhood graph?

Branch name with "Vignette 2.0" is also a good option :)

## Calculate Moran's I score for each gene
Expand Down